Fault detected in 5800X3D multiplier unit
Fault detected in 5800X3D multiplier unit
Hey everyone, I recently upgraded my X570 system last week with a 5800X3D, RTX 4090, a 42-inch 4K OLED display. It’s been performing well overall. Recently, I’ve been dealing with some issues while playing long gaming sessions at home because of an injury that limits my mobility. The system has been struggling, especially during reboots and even when using UEFI settings.
I noticed the CPU is throttling significantly—down to about 5x its normal speed—which makes everything feel very slow. Initially, this persisted after reboots, and even after resetting the CMOS, it didn’t improve. However, a complete shutdown resolved the problem. I took screenshots with HWinfo and managed to fix it by powering down the system instead of rebooting.
I’m confident my BIOS and chipset drivers are up to date (though I’m not sure). The issue seems related to undervolting; I’ve tried reducing voltage by 50mV via TPU2, but temperatures didn’t seem to be a factor. The first incident happened after an hour of gaming, but it was smooth the next day. Now, with another extended session this evening, it’s reoccurring.
My question is: if the undervolt adjustment didn’t help, would the system just crash completely? Any suggestions on what else could be causing this? Thanks!
How did you manage to achieve 5800X3D at 300W? That doesn’t add up. Are the measurements accurate? Under maximum workloads like Linpack Extreme, the CPU typically draws 120W to 140W. Cooling it down to 44°C with a 300W load seems impossible. I’m sure your HWinfo data is faulty—try reinstalling or re-downloading it.